- Email: [email protected]
Stability of the unlinked Latitude total elbow prosthesis: A biomechanical in vitro analysis
Clinical Biomechanics 28 (2013) 502–508
Contents lists available at SciVerse ScienceDirect
Clinical Biomechanics journal homepage: www.elsevier.com/locate/clinbiomech
Stability of the unlinked Latitude total elbow prosthesis: A biomechanical in vitro analysis Marc L. Wagener a,⁎, Maarten J. De Vos b, Jan C.M. Hendriks c, Denise Eygendaal d, Nico Verdonschot a, e a
Department of Orthopaedics, St. Radboud University Hospital, Pb 9101, 6500HB, Nijmegen, The Netherlands Department of Orthopaedics, TerGooi Hospital, Pb 10016, 1201DA, Hilversum, The Netherlands Department of Epidemiology, Biostatistics and Health Technology, Pb 9101, 6500HB, Nijmegen, The Netherlands d Department of Orthopaedics, Amphia Hospital, Upper Limb Unit, Pb 90157, 4800RL, Breda, The Netherlands e Laboratory for Biomechanical Engineering, University of Twente, 7500AE, Enschede, The Netherlands b c
a r t i c l e
i n f o
Article history: Received 27 January 2013 Accepted 29 April 2013 Keywords: Elbow Arthroplasty Stability Modularity
a b s t r a c t Background: The purpose of this study is to assess the valgus and varus laxity of the unlinked version of the Latitude total elbow prosthesis and the effects of radial head preservation or replacement. Methods: Biomechanical analysis of the valgus and varus laxity of the unlinked Latitude was performed in fourteen upper limb specimens in the following conditions: (1) native elbow, (2) native elbow after the surgical approach and closing all layers again, (3) elbow with humeral and ulnar component implanted, unlinked, with the native radial head preserved, (4) elbow with humeral and ulnar component implanted, unlinked, with the native radial excised, (5) elbow with humeral, ulnar, and radial head component implanted. Findings: After implantation of the Latitude total elbow prosthesis both the valgus and varus laxity slightly increase from mid to maximal flexion when compared to the native elbow after surgical approach. The unlinked Latitude total elbow prosthesis provides both valgus and varus stability in elbows with intact ligamentous constraints. With intact ligamentous constraints the radial head component only slightly contributes to the stability of the elbow after implantation of the unlinked Latitude total elbow prosthesis. Interpretation: The unlinked Latitude total elbow prosthesis provides both valgus and varus stability in elbows with intact ligamentous constraints. The radial head component contributes only slightly to the stability. © 2013 Elsevier Ltd. All rights reserved.
1. Introduction In 1970 Dee and Sweetnam described two cases in which a “fully” constrained elbow prosthesis was placed. In the following 40 years the total elbow prosthesis has developed substantially. Nowadays total elbow arthroplasty is the gold standard in the treatment of advanced arthritis of the joint caused by rheumatoid arthritis, primary osteoarthritis, or (post-)traumatic deformities. Both mechanically linked prostheses and unlinked prostheses are available. Most total elbow prostheses consist solely of a humeral and ulnar component. In linked prostheses the radial head can be retained, but it does not contribute to the stability of the elbow. Most unlinked prostheses do not offer an articulating surface that resembles the capitellum. In those prostheses the radial head is excised, but unlinked elbow prostheses with a capitellum and the option of a radial head component are also available. In the native elbow the role of the radial head in axial load bearing and stability has been re-emphasized. The radial head contributes to
⁎ Corresponding author. E-mail addresses: [email protected] (M.L. Wagener), [email protected] (M.J. De Vos), [email protected] (J.C.M. Hendriks), [email protected] (D. Eygendaal), [email protected] (N. Verdonschot). 0268-0033/$ – see front matter © 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.clinbiomech.2013.04.011
the stability of the elbow when valgus stress is applied, especially in elbows with insufficiency of the medial collateral ligament (MCL) (Beingessner et al., 2004; Jensen et al., 2005; Morrey et al., 1991). Load transmission across the radial head depends on the amount of flexion of the elbow and can be up to 60% (Morrey and An, 2005; Morrey et al., 1988). Previous studies show that the surrounding ligaments mainly maintain the stability of the elbow with an unlinked elbow prosthesis implanted. The intrinsic stability of the unlinked prosthesis (Kamineni et al., 2005) and a radial head component, if implanted, also contribute to the stability (Inagaki et al., 2002; Ramsey et al., 2003). In linked elbow prostheses on the other hand, the stability is guaranteed by the design of the prosthesis. Due to wear of the articular surface of the unbalanced linked elbow prosthesis, instability might occur. The Latitude (Tornier, Stafford, TX) total elbow prosthesis offers prosthetic modularity and convertibility. Intra-operatively the decision can be made whether the native radial head is retained, resected, or replaced by a bipolar radial head component. The prosthesis can be placed in an unlinked or a linked version (Szekeres and King, 2006). The condition of the stabilizing ligaments and the amount of bone loss is assessed during surgery. These findings, in relation to the demand of the patient, will finally support the surgeons' decision in which modularity the prosthesis is placed. However, how different modularities affect biomechanical stability of the elbow joint has not been analyzed yet.
M.L. Wagener et al. / Clinical Biomechanics 28 (2013) 502–508
Therefore, the purpose of this study was to assess the valgus and varus laxity of the unlinked version of the Latitude total elbow prosthesis relative to the intact elbow joint. Modularities that were assessed were the unlinked implant with (1) the native radial head preserved, (2) the native radial head excised, and (3) the native radial head replaced by a radial component.
2. Methods Fourteen fresh-frozen upper limb specimens from 7 donors (3 men, 4 women) who were 50 to 93 years old (mean, 71 years old) were available. All frozen arms were dissected through the humerus proximal of the deltoid insertion. The frozen arms were thawed overnight at room temperature. All specimens were macroscopically assessed and all osseous structures were radiologically analyzed before including them in the study. External macroscopical inspection and radiographic analysis of all arms showed no abnormalities of upper arm, elbow joint, forearm, and wrist. No signs of previous surgery were found in any of the arms. The skin and subcutaneous tissue of the proximal half of the upper-arm were removed. All muscles of the upper-arm were removed except for the biceps, triceps and brachialis muscle. These three muscles were all detached from the humerus and the surrounding soft tissue. The triceps and biceps muscles were dissected at the level of the most proximal point of the insertion of the brachialis muscle, leaving all muscles with the same length. The wrist was disarticulated, leaving the TFCC intact. The prepared specimens were mounted on to a specially designed testing apparatus. The apparatus allowed the humerus to be rotated along its longitudinal axis. This allowed either a valgus or a varus forces to be applied to the forearm using a weight that was hanged from a pin in the distal ulna. When the forearm was turned downwards no varus or valgus force was applied to the elbow (Fig. 1). In previous studies a 0.75 N moment applied to the forearm had proven to result in measurable varus and valgus deviations without resulting in damage to the collateral ligaments of the elbow (Jensen et al., 1999, 2005; O'Driscoll et al., 1992; Olsen et al., 1996). The weight that was needed to apply 0.75 N force was determined depending to the length of the forearm. Forces of 20 N, 10 N, and 10 N were applied to the triceps, biceps, and brachialis muscle, respectively. This imitates active contraction of the biceps, triceps and brachialis muscle, which results in increased stability of the elbow and this provides a better representation of the normal situation (King et al., 1993, 1994; O'Driscoll et al., 1992). A pin was inserted into the distal ulna and into the distal radius. The forearm was fixed in neutral rotation by a specially designed apparatus. The flexion and extension of the elbow was realized manually. Flexion and extension was always performed to the maximum that was possible in each condition. One of the researchers used a rod to push against the pin that was inserted distally into the ulna. The rod was held perpendicular to the flexion-extension direction of the elbow and no friction between the ulnar pin and the rod was noticed. During the testing the specimens were kept in optimal condition by keeping them moist with a 0.9% saline solution.
503
2.1. Surgical technique Placement of the prosthesis was performed as described by Gramstad et al. (2005) and Szekeres and King (2006). The management of the triceps was done in a triceps-tongue technique. The annular ligament was released from the ulna by a small bone chip that could easily be re-fixated with a trans-osseous suture, as described in the Wrightington approach (Stanley et al., 2006). Gramstad et al. (2005) describe a sharp release of both the medial and lateral collateral ligaments for placement of the Latitude elbow prosthesis. Repeated re-fixation of sharply released ligaments in a testing setting is likely to result in laxity of the elbow due to deterioration of the ligament. Therefore, release of the medial collateral ligament was performed by an osteotomy of the medial epicondyle. The epicondyle was re-fixated with trans-osseous sutures. Release of the stabilizing structures on the lateral side of the elbow is not necessary for optimal placement of the prosthesis, except for the earlier mentioned release of the annular ligament.
2.2. Motion tracking An electromagnetic motion tracking system (3SPACE Fastrak, Polhemus, Colchester, VT) allowed for measurement of the 3-dimensional position and orientation of a sensor in relation to a source. The source was fixed rigidly onto the testing apparatus and the sensor was mounted rigidly onto the ulnar pin. The location and orientation of the sensor was recorded using continuous data acquisition. Flexion of the elbow, varus and valgus laxity, and rotation of the ulna were recorded. The accuracy of this system has been reported to be 0.5° (An et al., 1988; Morrey et al., 1991). Because this system worked with a magnetic field all materials used in the close proximity of the specimens during the testing were made of materials that had no influence on the magnetic field.
2.3. Tested conditions All conditions were measured with the forearm in neutral position. The experiments were performed under the following conditions: (1) Native elbow. (2) Native elbow after the surgical approach and closing all layers again. This gave information on the influence of the surgical procedure on the stability of the elbow. The following conditions all include closing of all layers: (3) Elbow with humeral and ulnar component implanted, unlinked, with the native radial head preserved. (4) Elbow with humeral and ulnar component implanted, unlinked, with the native radial head excised. (5) Elbow with humeral, ulnar, and radial head component implanted.
Fig. 1. Specially designed testing apparatus in which the humerus can be rotated along its longitudinal axis. In the picture the forearm is in full extension. The weight hanging from the rod in the ulna applies a varus force to the forearm. A indicates a sensor and a specially designed apparatus that achieved fixed rotation of the forearm.
M.L. Wagener et al. / Clinical Biomechanics 28 (2013) 502–508
Instability (degrees)
504
Flexion (degrees)
Flexion (degrees)
Fig. 2. The observed individual instability profiles of all seven left native elbows with a varus force applied to the forearm. The figure on the left shows the profiles during extension of the elbow, the figure on the right shows the profiles during flexion of the elbow. Notice the typical pattern of a fourth-degree polynomial.
2.4. Valgus and varus laxity Throughout this study the valgus and varus laxity of the elbow at a certain degree of flexion are defined as follows: The valgus-or-varus angle of the forearm in space at a certain degree of flexion with a force applied to the forearm, minus the valgus-or-varus angle of the forearm when no force is applied to the forearm at the same degree of flexion. Hence, the laxity indicates the amount of extra valgus-or-varus angulation due the external force applied. The laxity profile is defined as the valgus or varus laxity as a function of the degrees of elbow flexion. The valgus and varus laxity of the elbow during each flexion-extension cycle was determined at every five degrees of flexion. 3. Statistical methods For each tested condition (1 through 5) a full flexion-extension cycle was performed three times, from which an average laxity profile during extension and flexion of the elbow was calculated. Close inspection of these average profiles showed, at first, that the results during extension and flexion of the elbow were nearly identical. Therefore the mean of the laxity during extension and flexion of the elbow at each flexion angle was used for analyses. During testing all elbows had been flexed and extended to the maximal of flexion-extension range, it appeared though that not all elbows had been flexed to zero and to 140 degrees of flexion, which
led to erroneous laxity patterns in these regions of elbow flexion. We therefore decided to limit the analyses to the flexion range between 10 and 130 degrees. Additionally, it became apparent that the laxity profiles showed a typical pattern starting with small values at extension, to larger values at around 20–40 degrees of flexion to smaller values again at higher flexion angles. A linear mixed model (Verbeke and Molenberghs, 1997) was used to fit a fourth-degree polynomial depending on flexion angle to the individual laxity profiles. The dependent variable was either the valgus or the varus laxity, respectively. The independent class variable was the test condition (5 different test conditions; see above). Also, the interaction terms between the test condition and the regression variables were included in the model. The intercept and the regression coefficients of flexion were treated as random effects. This way, differences between individual profiles are optimal allowed. The likelihood ratio test showed that when polynomials were used with either higher or lower degree than four, the fit was statistical significant decreased. The estimated regression parameters with standard errors were used to calculate the mean laxity-profiles with 95% confidence band for each condition. Furthermore, the differences of laxity-profiles with 95% confidence band between two conditions are visualized in Fig. 4. The estimated mean difference, with the 95% C.I., between the valgus and varus laxity, respectively, were calculated using a linear mixed model between the following conditions: (A) Elbow after surgical approach compared to the native elbow. (B) Unlinked elbow prosthesis with native radial head compared to the elbow after surgical approach. (C) Unlinked elbow prosthesis with radial head excised compared to the unlinked elbow prosthesis with native radial head. (D) Unlinked elbow prosthesis compared to the elbow after surgical approach. (E) Unlinked elbow prosthesis with radial head component compared to the unlinked elbow prosthesis with radial head excised. (F) Unlinked elbow prosthesis with radial head component compared to the elbow after surgical approach. Statistical analyses were performed using SAS 9.2 for Windows. 4. Results 4.1. Complications
Fig. 3. The observed and estimated individual laxity profiles of four randomly selected left native elbows with a varus force applied to the forearm. The stars indicate the observed values and the lines indicate the estimated profiles, using a fourth-degree polynomial in a linear mixed model.
During the testing one complication occurred. One of the elbows with the humeral and ulnar component in place with the native radial head preserved, showed recurrent dislocations during the testing of
M.L. Wagener et al. / Clinical Biomechanics 28 (2013) 502–508
505
Fig. 4. The estimated mean difference between the laxity profiles of two conditions in the direction of valgus and the direction of varus, respectively, using a linear mixed model. The dashed lines indicate the 95% confidence bands. In all figures increase of laxity in the direction of valgus is shown in the bottom part of the figure, and increase of laxity in the direction of varus is shown in the upper part of the figure. For example, figure A shows that the both the valgus and varus laxity of the native elbow is slightly lower (i.e. more stable) compared to after surgical approach. Figure B, D and E show that the both the valgus and varus laxity increase after implantation of the prosthesis.
the valgus stability. Near full extension the elbow functioned comparably to the other elbows used for testing. From approximately 70° of flexion dislocation of the elbow occurred. In the calculation of the estimated value of the valgus laxity this elbow was left out. The cause for the dislocation was not clear. Since the elbow showed normal kinematics after excision of the radial head, it was again included for further testing and analysis.
4.2. Stability of the elbow Fig. 3 shows the observed and estimated varus laxity profiles of four randomly selected elbows in the same condition (native elbow). This figure shows that the observed data are in close agreement with the best-estimated fourth-degree polynomial, using a linear mixed model. The within subject (residual) standard deviation of the final model
506
M.L. Wagener et al. / Clinical Biomechanics 28 (2013) 502–508
was 0.12 (valgus) and 0.10 (varus) and the goodness of fit (R2) was 97% (valgus) and 98% (varus). The observed data of the elbows with the prosthesis implanted also showed the typical pattern of a fourthdegree polynomial. Furthermore, it shows that this model is sufficient flexible to fit the various instability patterns (Fig. 2). Table 1 shows the estimated mean varus and valgus laxity by degree of flexion for each tested condition. In the native elbow the mean valgus and varus laxity decrease towards maximal flexion, as has previously been described in literature (Jensen et al., 1999; Morrey et al., 1991). A decrease of mean valgus and varus laxity was towards maximal flexion was also found after surgical exposure. For all conditions with the prosthesis implanted, both the mean varus and valgus laxity increased towards maximal flexion. Fig. 4 shows the estimated difference between the varus and valgus laxity of two conditions. The differences between the conditions are more extensively described below. (A) Elbow after surgical exposure versus native elbow Fig. 4A shows increase in valgus laxity of the elbow due to the surgical exposure is found in the complete flexion/extension cycle (P = 0.005), with a maximal increase of 2.4° (95% C.I. 0.9°; 3.9°) at 110° of flexion. Increase in varus laxity due to the surgical exposure is only seen close to maximal extension of the elbow (P = 0.003), with a maximum of 1.8° (95% C.I. 0.0°; 3.6°) at 10° of flexion. (B) Unlinked prosthesis with native radial head versus elbow after surgical exposure Both valgus and varus laxity increase when the elbow is further flexed 50° in the elbow with unlinked prosthesis Table 1 The estimated mean varus and valgus laxity (degree) with 95% confidence interval by flexion by condition, using a linear mixed model. Valgus
Varus
5. Discussion
Condition
Flexion (degree)
Mean (95% C.I.)
Mean (95% C.I.)
Native elbow
10 30 50 70 90 110 130 10 30 50 70 90 110 130 10 30 50 70 90 110 130 10 30 50 70 90 110 130 10 30 50 70 90 110 130
4.4 (5.4; 3.3) 4.0 (4.8; 3.3) 3.5 (4.1; 2.9) 3.0 (3.7; 2.4) 2.7 (3.5; 2.0) 2.7 (3.7; 1.6) 2.7 (4.6; 0.7) 6.4 (7.5; 5.4) 6.0 (6.7; 5.3) 5.5 (6.1; 4.9) 5.2 (5.8; 4.5) 5.1 (5.8; 4.3) 5.1 (6.1; 4.0) 4.8 (6.7; 2.9) 5.0 (7.3; 2.7) 6.7 (8.2; 5.3) 7.2 (8.4; 6.1) 8.0 (9.1; 6.9) 9.3 (10.7; 7.8) 10.2 (12.4; 7.8) 8.5 (13.2; 3.8) 6.9 (9.8; 4.0) 8.2 (10.3; 6.2) 9.0 (10.7; 7.3) 10.3 (11.9; 8.6) 12.0 (14.0; 9.9) 13.4 (16.2; 10.5) 12.7 (17.6; 7.8) 7.0 (9.9; 4.1) 7.9 (10.3; 5.5) 8.0 (10.3; 5.8) 8.6 (10.8; 6.3) 9.7 (12.1; 7.3) 10.8 (13.7; 7.9) 9.8 (14.1; 5.5)
4.5 (3.2; 5.2 (4.3; 4.5 (3.7; 3.5 (2.8; 2.9 (2.1; 2.9 (1.7; 3.0 (0.6; 6.2 (5.0; 6.4 (5.5; 5.1 (4.4; 3.9 (3.1; 3.3 (2.4; 3.4 (2.2; 3.6 (1.2; 5.2 (3.5; 6.5 (5.4; 6.6 (5.6; 6.3 (5.4; 6.3 (5.2; 6.7 (5.0; 7.3 (4.0; 7.1 (5.2; 8.1 (6.8; 7.7 (6.6; 7.2 (6.1; 7.1 (5.8; 7.5 (5.7; 7.8 (4.6; 5.4 (4.0; 6.3 (5.1; 6.4 (5.3; 6.5 (5.4; 6.8 (5.7; 7.3 (5.9; 7.4 (4.7;
After surgical exposure
Unlinked, native radial head
Unlinked, radial head excised
Unlinked, with radial head component
implanted with the native radial head retained (Fig. 4B). For the difference in valgus laxity between the two conditions P b 0.001, for the difference in varus laxity P b 0.001. (C and D) Unlinked prosthesis after excision of the radial head compared to the unlinked prosthesis with native radial head, and compared to the elbow after surgical exposure. Fig. 4C shows that resection of the radial head results in increase in varus laxity (P = 0.074). The valgus laxity increases more the varus laxity, with maximal increase in the direction of full flexion (P = 0.029). Increase of varus and valgus laxity was mainly found in mid-flexion and maximal-flexion when compared to the laxity after surgical exposure (Valgus P b 0.001; varus P b 0.001) (Fig. 4D). (E and F) Unlinked prosthesis with radial head prosthesis compared to the unlinked prosthesis with radial head excised, and compared to the elbow after surgical exposure. When the unlinked prosthesis with radial head prosthesis implanted was compared to the elbow after surgical exposure an increase of both varus and valgus laxity was found (Valgus P = 0.002; varus P b 00.1). An increase of both valgus and varus laxity from mid-flexion to maximal-flexion was seen. A maximum increase of both valgus and varus laxity was found at 110° of flexion. (Fig. 4F). Implantation of the radial head prosthesis results in improved valgus and varus laxity when compared to the unlinked prosthesis with the radial head excised (Fig. 4E). Valgus laxity improves in mid flexion and maximal flexion (P = 0.629). Varus laxity improves near complete extension (P = 0.390). Both instability in the direction of valgus and varus are restored to the amount of instability that was found with the native radial preserved (Table 1).
5.7) 6.1) 5.3) 4.3) 3.8) 4.1) 5.4) 7.5) 7.3) 5.9) 4.6) 4.2) 4.7) 5.9) 6.9) 7.6) 7.5) 7.3) 7.4) 8.4) 10.6) 8.9) 9.4) 8.8) 8.3) 8.4) 9.3) 11.0) 6.9) 7.4) 7.5) 7.6) 8.0) 8.7) 10.1)
This study shows that the stability of the elbow after implantation of the unlinked Latitude total elbow prosthesis is altered when compared to the native elbow. In all tested conditions both the valgus and varus laxity increase from mid to maximal flexion when compared to the native elbow after surgical approach. Maximal increase in both valgus and varus laxity is seen near full flexion in the condition after resection of the radial head. In this condition maximal increase in valgus laxity is approximately 7° and maximal increase in varus laxity is approximately 4° when compared to the condition after the surgical exposure. Crucial in the placement of the Latitude prosthesis is the reproduction of the center of rotation axis (Gramstad et al., 2005). For this purpose meticulous exposure of all ligamentous and bony landmarks is essential. This was achieved by performing an osteotomy of the medial epicondyle and a bony release of the annular ligament. The medial epicondylar osteotomy allowed optimal and anatomical re-fixation of the stabilizing structures on the medial side of the elbow. Optimal fixation of the medial epicondyle was achieved with the trans-osseous sutures in all specimens. The stability tests that were performed before and after surgical exposure showed that the surgical approach had only minimal clinical influence on the elbow kinematics, although it was found to be statistically significant. The changes in valgus and varus stability of the elbow after implantation of the Latitude prosthesis could therefore be fully contributed to the prosthesis design, e.g. the intrinsic constraint of the prosthesis, and the orientation of the prosthesis in relation to the bony landmarks and the soft-tissue constraints. The medial ligamentous complex is the primary stabilizer of the elbow during valgus loading. The radial head is known to be a secondary stabilizer during valgus loading (Beingessner et al., 2004; Jensen et al., 1999, 2005; Morrey et al., 1991). In this study, excision of the radial head resulted in minimal increase of valgus laxity. A critical
M.L. Wagener et al. / Clinical Biomechanics 28 (2013) 502–508
note must be made that all used elbows had intact ligamentous constraints. Elbows with rheumatoid arthritis or posttraumatic deformities often show insufficiency of the collateral ligaments, resulting in diminished stability of the elbow joint. Since elbow arthroplasty is often performed in patients with rheumatoid arthritis, who often have insufficiency of the ligamentous constraints, it can be expected that the radial head will often contribute more to the stability of the elbow during valgus loads than is demonstrated in this study. Furthermore it must be noticed that the radial head can also contribute more to the stability of the elbow when healing of the medial ligamentous complex is impaired after the surgical exposure, or when ligamentous laxity occurs in time after deterioration of the quality of the medial ligamentous complex. In the native elbow excision of the radial head also results in increased instability during varus loading of the elbow. This increase in varus instability is mainly seen near full extension (Jensen et al., 1999). The increase of varus instability can be explained by slacking of the lateral ligament complex after excision of the radial head. In this study a similar, slight, increase in varus instability is seen after excision of the native radial head near full extension. Implantation of the radial head component results in restoration of the stability of the elbow as seen before excision of the native radial head. Valgus stability is improved from mid-flexion to full-flexion when compared to the condition with the radial head excised. Varus stability is improved near full extension. Moreover, restoration of the correct height of the radial head implant is of utmost importance to obtain stability. The correct height of the radial head component is difficult to determine during surgery (Doornberg et al., 2006; Frank et al., 2009; Glabbeek van et al., 2005). Other unlinked elbow prostheses that have the option of implantation of radial head component are the Wright Sorbie-Questor (Wright Medical Technology, Mississauga, ON, Canada) and the Pritchard-ERS (Depuy, Warsaw, IN). The contribution of a monoblock radial head component to the valgus stability of the Wright Sorbie-Questor unlinked total elbow prosthesis was previously described (Inagaki et al., 2002). Contrary to our study they showed increased valgus stability from mid-flexion to full-extension when the radial head component was implanted. This can most likely be explained by a difference in prosthesis design and intrinsic constraint. The kinematic characteristics of the Pritchard-ERS total elbow prosthesis was described by Ramsey et al. (2003). A similar valgus laxity was found as seen in the Latitude prosthesis. The varus laxity of the Pritchard-ERS with radial head component implanted is on the other hand, considerably greater than the varus laxity found in our study. Excision of the radial head resulted in considerably more increase of both valgus and varus laxity than was found with the Latitude total elbow prosthesis. Other studies, in which the contribution of the radial head component to the stability of total elbow prosthesis, e.g. the Mayo Elbow and the AHSC-Volz elbow prosthesis, have to our knowledge not been performed. In the study of King et al. (1994) the stability of the unlinked Capitellocondylar (Johnson and Johnson Orthopaedics, New Brunswick, New Jersey) total elbow prosthesis was analyzed. Since in this study the varus and valgus laxity were added up and presented as one mean maximum valgus-varus laxity for the seventeen used specimens, it is difficult to compare our data to the results of this study. A similar increase of valgus laxity in the direction of full flexion is seen in the figures as presented in their study. Clinically the valgus and varus instability as seen in this study would be scored as “Moderate instability,” according to the “the Mayo Clinic Performance Index for the Elbow” (Morrey and Adams, 1992). Great care should be taken though when translating these biomechanical testing results into clinical functional characteristics. The subjective stability of the elbow that is experienced is grossly dependent on patient demand and expectations. Furthermore the imitated active muscle loads as applied in this study do provide stability to
507
the elbow, but obviously do not match the activity dependent muscle loads in real life. Further research should be performed whether the unlinked version of the Latitude provides enough stability in elbows with insufficiency of collateral ligaments. It might very well be that in elbows with insufficiency of the medial and/or lateral stabilizing structures the unlinked version of the Latitude does not provide enough stability, and that the patient is better off when a linked version of this prosthesis is implanted. 6. Conclusion In conclusion we can state that the unlinked Latitude total elbow prosthesis provides both valgus and varus stability in elbows with intact ligamentous constraints, and that the radial head component only slightly contributes to the stability. Conflict of interest statement No financial remunerations were received by any of the authors. Acknowledgements The authors acknowledge the contributions of M. Barink, PhD, and A. Venner for their support and advice. Financial resources were supplied by Tornier (Stafford, TX) for the purchase of the upper limb specimens, the laboratorial costs, and the costs for the statistical support. The elbow prostheses used in the study were supplied by Tornier. Tornier had no influence on the study design, the collection, analysis, and interpretation of data, or in the writing of the manuscript. References An, K.N., Jacobsen, M.C., Berglund, L.J., Chao, E.Y., 1988. Application of a magnetic tracking device to kinesiologic studies. J. Biomech. 21, 613–620. Beingessner, D.M., Dunning, C.E., Gordon, K.D., Johnson, J.A., King, G.J., 2004. The effect of radial head excision and arthroplasty on elbow kinematics and stability. J. Bone Joint Surg. Am. 86-A, 1730–1739. Dee, R., Sweetnam, D.R., 1970. Total replacement arthroplasty of the elbow joint for rheumatoid arthritis: two cases. Proc. R. Soc. Med. 63, 653–655. Doornberg, J.N., Linzel, D.S., Zurakowski, D., Ring, D., 2006. Reference points for radial head prosthesis size. J. Hand Surg. Am. 31, 53–57. Frank, S.G., Grewal, R., Johnson, J., Faber, K.J., King, G.J., Athwal, G.S., 2009. Determination of correct implant size in radial head arthroplasty to avoid overlengthening. J. Bone Joint Surg. Am. 91, 1738–1746. Glabbeek van, F., Riet van, R.P., Baumfeld, J.A., Neale, P., O'Driscoll, S.W., Morrey, B.F., 2005. The kinematic importance of radial neck length in radial head placement. Med. Eng. Phys. 27, 336–342. Gramstad, G.D., King, G.J., O'Driscoll, S.W., Yamaguchi, K., 2005. Elbow arthroplasty using a convertible implant. Tech. Hand Up Extrem. Surg. 9, 153–163. Inagaki, K., O'Driscoll, S.W., Neale, P.G., Uchiyama, E., Morrey, B.F., An, K.N., 2002. Importance of a radial head component in Sorbie unlinked total elbow arthroplasty. Clin. Orthop. Relat. Res. 123–131. Jensen, S.L., Olsen, B.S., Sojbjerg, J.O., 1999. Elbow joint kinematics after excision of the radial head. J. Shoulder Elbow Surg. 8, 238–241. Jensen, S.L., Olsen, B.S., Tyrdal, S., Sojbjerg, J.O., Sneppen, O., 2005. Elbow joint laxity after experimental radial head excision and lateral collateral ligament rupture: efficacy of prosthetic replacement and ligament repair. J. Shoulder Elbow Surg. 14, 78–84. Kamineni, S., O'Driscoll, S.W., Urban, M., Garg, A., Berglund, L.J., Morrey, B.F., An, K.N., 2005. Intrinsic constraint of unlinked total elbow replacements—the ulnotrochlear joint. J. Bone Joint Surg. Am. 87, 2019–2027. King, G.J., Glauser, S.J., Westreich, A., Morrey, B.F., An, K.N., 1993. In vitro stability of an unconstrained total elbow prosthesis. Influence of axial loading and joint flexion angle. J. Arthroplasty 8, 291–298. King, G.J., Itoi, E., Niebur, G.L., Morrey, B.F., An, K.N., 1994. Motion and laxity of the capitellocondylar total elbow prosthesis. J. Bone Joint Surg. Am. 76, 1000–1008. Morrey, B.F., Adams, R.A., 1992. Semiconstrained arthroplasty for the treatment of rheumatoid arthritis of the elbow. J. Bone Joint Surg. Am. 74, 479–490. Morrey, B.F., An, K.N., 2005. Stability of the elbow: osseous constraints. J. Shoulder Elbow Surg. 14, 174S–178S. Morrey, B.F., An, K.N., Stormont, T.J., 1988. Force transmission through the radial head. J. Bone Joint Surg. Am. 70, 250–256. Morrey, B.F., Tanaka, S., An, K.N., 1991. Valgus stability of the elbow. A definition of primary and secondary constraints. Clin. Orthop. Relat. Res. 187–195.
508
M.L. Wagener et al. / Clinical Biomechanics 28 (2013) 502–508
O'Driscoll, S.W., An, K.N., Korinek, S., Morrey, B.F., 1992. Kinematics of semi-constrained total elbow arthroplasty. J. Bone Joint Surg. Br. 74, 297–299. Olsen, B.S., Sojbjerg, J.O., Dalstra, M., Sneppen, O., 1996. Kinematics of the lateral ligamentous constraints of the elbow joint. J. Shoulder Elbow Surg. 5, 333–341. Ramsey, M., Neale, P., Morrey, B.F., O'Driscoll, S.W., An, K.N., 2003. Kinematics and functional characteristics of the Pritchard ERS unlinked total elbow arthroplasty. J. Shoulder Elbow Surg. 12, 385–390.
Stanley, J., Penn, D., Wasseem, M., 2006. Exposure of the head of the radius using the Wrightington approach. J. Bone Joint Surg. Br. 88, 1178–1182. Szekeres, M., King, G.J., 2006. Total elbow arthroplasty. J. Hand Ther. 19, 245–253. Verbeke, G., Molenberghs, G., 1997. Linear Mixed Models in Practice: An SAS-Oriented Approach. Springer, New York.
Contents lists available at SciVerse ScienceDirect
Clinical Biomechanics journal homepage: www.elsevier.com/locate/clinbiomech
Stability of the unlinked Latitude total elbow prosthesis: A biomechanical in vitro analysis Marc L. Wagener a,⁎, Maarten J. De Vos b, Jan C.M. Hendriks c, Denise Eygendaal d, Nico Verdonschot a, e a
Department of Orthopaedics, St. Radboud University Hospital, Pb 9101, 6500HB, Nijmegen, The Netherlands Department of Orthopaedics, TerGooi Hospital, Pb 10016, 1201DA, Hilversum, The Netherlands Department of Epidemiology, Biostatistics and Health Technology, Pb 9101, 6500HB, Nijmegen, The Netherlands d Department of Orthopaedics, Amphia Hospital, Upper Limb Unit, Pb 90157, 4800RL, Breda, The Netherlands e Laboratory for Biomechanical Engineering, University of Twente, 7500AE, Enschede, The Netherlands b c
a r t i c l e
i n f o
Article history: Received 27 January 2013 Accepted 29 April 2013 Keywords: Elbow Arthroplasty Stability Modularity
a b s t r a c t Background: The purpose of this study is to assess the valgus and varus laxity of the unlinked version of the Latitude total elbow prosthesis and the effects of radial head preservation or replacement. Methods: Biomechanical analysis of the valgus and varus laxity of the unlinked Latitude was performed in fourteen upper limb specimens in the following conditions: (1) native elbow, (2) native elbow after the surgical approach and closing all layers again, (3) elbow with humeral and ulnar component implanted, unlinked, with the native radial head preserved, (4) elbow with humeral and ulnar component implanted, unlinked, with the native radial excised, (5) elbow with humeral, ulnar, and radial head component implanted. Findings: After implantation of the Latitude total elbow prosthesis both the valgus and varus laxity slightly increase from mid to maximal flexion when compared to the native elbow after surgical approach. The unlinked Latitude total elbow prosthesis provides both valgus and varus stability in elbows with intact ligamentous constraints. With intact ligamentous constraints the radial head component only slightly contributes to the stability of the elbow after implantation of the unlinked Latitude total elbow prosthesis. Interpretation: The unlinked Latitude total elbow prosthesis provides both valgus and varus stability in elbows with intact ligamentous constraints. The radial head component contributes only slightly to the stability. © 2013 Elsevier Ltd. All rights reserved.
1. Introduction In 1970 Dee and Sweetnam described two cases in which a “fully” constrained elbow prosthesis was placed. In the following 40 years the total elbow prosthesis has developed substantially. Nowadays total elbow arthroplasty is the gold standard in the treatment of advanced arthritis of the joint caused by rheumatoid arthritis, primary osteoarthritis, or (post-)traumatic deformities. Both mechanically linked prostheses and unlinked prostheses are available. Most total elbow prostheses consist solely of a humeral and ulnar component. In linked prostheses the radial head can be retained, but it does not contribute to the stability of the elbow. Most unlinked prostheses do not offer an articulating surface that resembles the capitellum. In those prostheses the radial head is excised, but unlinked elbow prostheses with a capitellum and the option of a radial head component are also available. In the native elbow the role of the radial head in axial load bearing and stability has been re-emphasized. The radial head contributes to
⁎ Corresponding author. E-mail addresses: [email protected] (M.L. Wagener), [email protected] (M.J. De Vos), [email protected] (J.C.M. Hendriks), [email protected] (D. Eygendaal), [email protected] (N. Verdonschot). 0268-0033/$ – see front matter © 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.clinbiomech.2013.04.011
the stability of the elbow when valgus stress is applied, especially in elbows with insufficiency of the medial collateral ligament (MCL) (Beingessner et al., 2004; Jensen et al., 2005; Morrey et al., 1991). Load transmission across the radial head depends on the amount of flexion of the elbow and can be up to 60% (Morrey and An, 2005; Morrey et al., 1988). Previous studies show that the surrounding ligaments mainly maintain the stability of the elbow with an unlinked elbow prosthesis implanted. The intrinsic stability of the unlinked prosthesis (Kamineni et al., 2005) and a radial head component, if implanted, also contribute to the stability (Inagaki et al., 2002; Ramsey et al., 2003). In linked elbow prostheses on the other hand, the stability is guaranteed by the design of the prosthesis. Due to wear of the articular surface of the unbalanced linked elbow prosthesis, instability might occur. The Latitude (Tornier, Stafford, TX) total elbow prosthesis offers prosthetic modularity and convertibility. Intra-operatively the decision can be made whether the native radial head is retained, resected, or replaced by a bipolar radial head component. The prosthesis can be placed in an unlinked or a linked version (Szekeres and King, 2006). The condition of the stabilizing ligaments and the amount of bone loss is assessed during surgery. These findings, in relation to the demand of the patient, will finally support the surgeons' decision in which modularity the prosthesis is placed. However, how different modularities affect biomechanical stability of the elbow joint has not been analyzed yet.
M.L. Wagener et al. / Clinical Biomechanics 28 (2013) 502–508
Therefore, the purpose of this study was to assess the valgus and varus laxity of the unlinked version of the Latitude total elbow prosthesis relative to the intact elbow joint. Modularities that were assessed were the unlinked implant with (1) the native radial head preserved, (2) the native radial head excised, and (3) the native radial head replaced by a radial component.
2. Methods Fourteen fresh-frozen upper limb specimens from 7 donors (3 men, 4 women) who were 50 to 93 years old (mean, 71 years old) were available. All frozen arms were dissected through the humerus proximal of the deltoid insertion. The frozen arms were thawed overnight at room temperature. All specimens were macroscopically assessed and all osseous structures were radiologically analyzed before including them in the study. External macroscopical inspection and radiographic analysis of all arms showed no abnormalities of upper arm, elbow joint, forearm, and wrist. No signs of previous surgery were found in any of the arms. The skin and subcutaneous tissue of the proximal half of the upper-arm were removed. All muscles of the upper-arm were removed except for the biceps, triceps and brachialis muscle. These three muscles were all detached from the humerus and the surrounding soft tissue. The triceps and biceps muscles were dissected at the level of the most proximal point of the insertion of the brachialis muscle, leaving all muscles with the same length. The wrist was disarticulated, leaving the TFCC intact. The prepared specimens were mounted on to a specially designed testing apparatus. The apparatus allowed the humerus to be rotated along its longitudinal axis. This allowed either a valgus or a varus forces to be applied to the forearm using a weight that was hanged from a pin in the distal ulna. When the forearm was turned downwards no varus or valgus force was applied to the elbow (Fig. 1). In previous studies a 0.75 N moment applied to the forearm had proven to result in measurable varus and valgus deviations without resulting in damage to the collateral ligaments of the elbow (Jensen et al., 1999, 2005; O'Driscoll et al., 1992; Olsen et al., 1996). The weight that was needed to apply 0.75 N force was determined depending to the length of the forearm. Forces of 20 N, 10 N, and 10 N were applied to the triceps, biceps, and brachialis muscle, respectively. This imitates active contraction of the biceps, triceps and brachialis muscle, which results in increased stability of the elbow and this provides a better representation of the normal situation (King et al., 1993, 1994; O'Driscoll et al., 1992). A pin was inserted into the distal ulna and into the distal radius. The forearm was fixed in neutral rotation by a specially designed apparatus. The flexion and extension of the elbow was realized manually. Flexion and extension was always performed to the maximum that was possible in each condition. One of the researchers used a rod to push against the pin that was inserted distally into the ulna. The rod was held perpendicular to the flexion-extension direction of the elbow and no friction between the ulnar pin and the rod was noticed. During the testing the specimens were kept in optimal condition by keeping them moist with a 0.9% saline solution.
503
2.1. Surgical technique Placement of the prosthesis was performed as described by Gramstad et al. (2005) and Szekeres and King (2006). The management of the triceps was done in a triceps-tongue technique. The annular ligament was released from the ulna by a small bone chip that could easily be re-fixated with a trans-osseous suture, as described in the Wrightington approach (Stanley et al., 2006). Gramstad et al. (2005) describe a sharp release of both the medial and lateral collateral ligaments for placement of the Latitude elbow prosthesis. Repeated re-fixation of sharply released ligaments in a testing setting is likely to result in laxity of the elbow due to deterioration of the ligament. Therefore, release of the medial collateral ligament was performed by an osteotomy of the medial epicondyle. The epicondyle was re-fixated with trans-osseous sutures. Release of the stabilizing structures on the lateral side of the elbow is not necessary for optimal placement of the prosthesis, except for the earlier mentioned release of the annular ligament.
2.2. Motion tracking An electromagnetic motion tracking system (3SPACE Fastrak, Polhemus, Colchester, VT) allowed for measurement of the 3-dimensional position and orientation of a sensor in relation to a source. The source was fixed rigidly onto the testing apparatus and the sensor was mounted rigidly onto the ulnar pin. The location and orientation of the sensor was recorded using continuous data acquisition. Flexion of the elbow, varus and valgus laxity, and rotation of the ulna were recorded. The accuracy of this system has been reported to be 0.5° (An et al., 1988; Morrey et al., 1991). Because this system worked with a magnetic field all materials used in the close proximity of the specimens during the testing were made of materials that had no influence on the magnetic field.
2.3. Tested conditions All conditions were measured with the forearm in neutral position. The experiments were performed under the following conditions: (1) Native elbow. (2) Native elbow after the surgical approach and closing all layers again. This gave information on the influence of the surgical procedure on the stability of the elbow. The following conditions all include closing of all layers: (3) Elbow with humeral and ulnar component implanted, unlinked, with the native radial head preserved. (4) Elbow with humeral and ulnar component implanted, unlinked, with the native radial head excised. (5) Elbow with humeral, ulnar, and radial head component implanted.
Fig. 1. Specially designed testing apparatus in which the humerus can be rotated along its longitudinal axis. In the picture the forearm is in full extension. The weight hanging from the rod in the ulna applies a varus force to the forearm. A indicates a sensor and a specially designed apparatus that achieved fixed rotation of the forearm.
M.L. Wagener et al. / Clinical Biomechanics 28 (2013) 502–508
Instability (degrees)
504
Flexion (degrees)
Flexion (degrees)
Fig. 2. The observed individual instability profiles of all seven left native elbows with a varus force applied to the forearm. The figure on the left shows the profiles during extension of the elbow, the figure on the right shows the profiles during flexion of the elbow. Notice the typical pattern of a fourth-degree polynomial.
2.4. Valgus and varus laxity Throughout this study the valgus and varus laxity of the elbow at a certain degree of flexion are defined as follows: The valgus-or-varus angle of the forearm in space at a certain degree of flexion with a force applied to the forearm, minus the valgus-or-varus angle of the forearm when no force is applied to the forearm at the same degree of flexion. Hence, the laxity indicates the amount of extra valgus-or-varus angulation due the external force applied. The laxity profile is defined as the valgus or varus laxity as a function of the degrees of elbow flexion. The valgus and varus laxity of the elbow during each flexion-extension cycle was determined at every five degrees of flexion. 3. Statistical methods For each tested condition (1 through 5) a full flexion-extension cycle was performed three times, from which an average laxity profile during extension and flexion of the elbow was calculated. Close inspection of these average profiles showed, at first, that the results during extension and flexion of the elbow were nearly identical. Therefore the mean of the laxity during extension and flexion of the elbow at each flexion angle was used for analyses. During testing all elbows had been flexed and extended to the maximal of flexion-extension range, it appeared though that not all elbows had been flexed to zero and to 140 degrees of flexion, which
led to erroneous laxity patterns in these regions of elbow flexion. We therefore decided to limit the analyses to the flexion range between 10 and 130 degrees. Additionally, it became apparent that the laxity profiles showed a typical pattern starting with small values at extension, to larger values at around 20–40 degrees of flexion to smaller values again at higher flexion angles. A linear mixed model (Verbeke and Molenberghs, 1997) was used to fit a fourth-degree polynomial depending on flexion angle to the individual laxity profiles. The dependent variable was either the valgus or the varus laxity, respectively. The independent class variable was the test condition (5 different test conditions; see above). Also, the interaction terms between the test condition and the regression variables were included in the model. The intercept and the regression coefficients of flexion were treated as random effects. This way, differences between individual profiles are optimal allowed. The likelihood ratio test showed that when polynomials were used with either higher or lower degree than four, the fit was statistical significant decreased. The estimated regression parameters with standard errors were used to calculate the mean laxity-profiles with 95% confidence band for each condition. Furthermore, the differences of laxity-profiles with 95% confidence band between two conditions are visualized in Fig. 4. The estimated mean difference, with the 95% C.I., between the valgus and varus laxity, respectively, were calculated using a linear mixed model between the following conditions: (A) Elbow after surgical approach compared to the native elbow. (B) Unlinked elbow prosthesis with native radial head compared to the elbow after surgical approach. (C) Unlinked elbow prosthesis with radial head excised compared to the unlinked elbow prosthesis with native radial head. (D) Unlinked elbow prosthesis compared to the elbow after surgical approach. (E) Unlinked elbow prosthesis with radial head component compared to the unlinked elbow prosthesis with radial head excised. (F) Unlinked elbow prosthesis with radial head component compared to the elbow after surgical approach. Statistical analyses were performed using SAS 9.2 for Windows. 4. Results 4.1. Complications
Fig. 3. The observed and estimated individual laxity profiles of four randomly selected left native elbows with a varus force applied to the forearm. The stars indicate the observed values and the lines indicate the estimated profiles, using a fourth-degree polynomial in a linear mixed model.
During the testing one complication occurred. One of the elbows with the humeral and ulnar component in place with the native radial head preserved, showed recurrent dislocations during the testing of
M.L. Wagener et al. / Clinical Biomechanics 28 (2013) 502–508
505
Fig. 4. The estimated mean difference between the laxity profiles of two conditions in the direction of valgus and the direction of varus, respectively, using a linear mixed model. The dashed lines indicate the 95% confidence bands. In all figures increase of laxity in the direction of valgus is shown in the bottom part of the figure, and increase of laxity in the direction of varus is shown in the upper part of the figure. For example, figure A shows that the both the valgus and varus laxity of the native elbow is slightly lower (i.e. more stable) compared to after surgical approach. Figure B, D and E show that the both the valgus and varus laxity increase after implantation of the prosthesis.
the valgus stability. Near full extension the elbow functioned comparably to the other elbows used for testing. From approximately 70° of flexion dislocation of the elbow occurred. In the calculation of the estimated value of the valgus laxity this elbow was left out. The cause for the dislocation was not clear. Since the elbow showed normal kinematics after excision of the radial head, it was again included for further testing and analysis.
4.2. Stability of the elbow Fig. 3 shows the observed and estimated varus laxity profiles of four randomly selected elbows in the same condition (native elbow). This figure shows that the observed data are in close agreement with the best-estimated fourth-degree polynomial, using a linear mixed model. The within subject (residual) standard deviation of the final model
506
M.L. Wagener et al. / Clinical Biomechanics 28 (2013) 502–508
was 0.12 (valgus) and 0.10 (varus) and the goodness of fit (R2) was 97% (valgus) and 98% (varus). The observed data of the elbows with the prosthesis implanted also showed the typical pattern of a fourthdegree polynomial. Furthermore, it shows that this model is sufficient flexible to fit the various instability patterns (Fig. 2). Table 1 shows the estimated mean varus and valgus laxity by degree of flexion for each tested condition. In the native elbow the mean valgus and varus laxity decrease towards maximal flexion, as has previously been described in literature (Jensen et al., 1999; Morrey et al., 1991). A decrease of mean valgus and varus laxity was towards maximal flexion was also found after surgical exposure. For all conditions with the prosthesis implanted, both the mean varus and valgus laxity increased towards maximal flexion. Fig. 4 shows the estimated difference between the varus and valgus laxity of two conditions. The differences between the conditions are more extensively described below. (A) Elbow after surgical exposure versus native elbow Fig. 4A shows increase in valgus laxity of the elbow due to the surgical exposure is found in the complete flexion/extension cycle (P = 0.005), with a maximal increase of 2.4° (95% C.I. 0.9°; 3.9°) at 110° of flexion. Increase in varus laxity due to the surgical exposure is only seen close to maximal extension of the elbow (P = 0.003), with a maximum of 1.8° (95% C.I. 0.0°; 3.6°) at 10° of flexion. (B) Unlinked prosthesis with native radial head versus elbow after surgical exposure Both valgus and varus laxity increase when the elbow is further flexed 50° in the elbow with unlinked prosthesis Table 1 The estimated mean varus and valgus laxity (degree) with 95% confidence interval by flexion by condition, using a linear mixed model. Valgus
Varus
5. Discussion
Condition
Flexion (degree)
Mean (95% C.I.)
Mean (95% C.I.)
Native elbow
10 30 50 70 90 110 130 10 30 50 70 90 110 130 10 30 50 70 90 110 130 10 30 50 70 90 110 130 10 30 50 70 90 110 130
4.4 (5.4; 3.3) 4.0 (4.8; 3.3) 3.5 (4.1; 2.9) 3.0 (3.7; 2.4) 2.7 (3.5; 2.0) 2.7 (3.7; 1.6) 2.7 (4.6; 0.7) 6.4 (7.5; 5.4) 6.0 (6.7; 5.3) 5.5 (6.1; 4.9) 5.2 (5.8; 4.5) 5.1 (5.8; 4.3) 5.1 (6.1; 4.0) 4.8 (6.7; 2.9) 5.0 (7.3; 2.7) 6.7 (8.2; 5.3) 7.2 (8.4; 6.1) 8.0 (9.1; 6.9) 9.3 (10.7; 7.8) 10.2 (12.4; 7.8) 8.5 (13.2; 3.8) 6.9 (9.8; 4.0) 8.2 (10.3; 6.2) 9.0 (10.7; 7.3) 10.3 (11.9; 8.6) 12.0 (14.0; 9.9) 13.4 (16.2; 10.5) 12.7 (17.6; 7.8) 7.0 (9.9; 4.1) 7.9 (10.3; 5.5) 8.0 (10.3; 5.8) 8.6 (10.8; 6.3) 9.7 (12.1; 7.3) 10.8 (13.7; 7.9) 9.8 (14.1; 5.5)
4.5 (3.2; 5.2 (4.3; 4.5 (3.7; 3.5 (2.8; 2.9 (2.1; 2.9 (1.7; 3.0 (0.6; 6.2 (5.0; 6.4 (5.5; 5.1 (4.4; 3.9 (3.1; 3.3 (2.4; 3.4 (2.2; 3.6 (1.2; 5.2 (3.5; 6.5 (5.4; 6.6 (5.6; 6.3 (5.4; 6.3 (5.2; 6.7 (5.0; 7.3 (4.0; 7.1 (5.2; 8.1 (6.8; 7.7 (6.6; 7.2 (6.1; 7.1 (5.8; 7.5 (5.7; 7.8 (4.6; 5.4 (4.0; 6.3 (5.1; 6.4 (5.3; 6.5 (5.4; 6.8 (5.7; 7.3 (5.9; 7.4 (4.7;
After surgical exposure
Unlinked, native radial head
Unlinked, radial head excised
Unlinked, with radial head component
implanted with the native radial head retained (Fig. 4B). For the difference in valgus laxity between the two conditions P b 0.001, for the difference in varus laxity P b 0.001. (C and D) Unlinked prosthesis after excision of the radial head compared to the unlinked prosthesis with native radial head, and compared to the elbow after surgical exposure. Fig. 4C shows that resection of the radial head results in increase in varus laxity (P = 0.074). The valgus laxity increases more the varus laxity, with maximal increase in the direction of full flexion (P = 0.029). Increase of varus and valgus laxity was mainly found in mid-flexion and maximal-flexion when compared to the laxity after surgical exposure (Valgus P b 0.001; varus P b 0.001) (Fig. 4D). (E and F) Unlinked prosthesis with radial head prosthesis compared to the unlinked prosthesis with radial head excised, and compared to the elbow after surgical exposure. When the unlinked prosthesis with radial head prosthesis implanted was compared to the elbow after surgical exposure an increase of both varus and valgus laxity was found (Valgus P = 0.002; varus P b 00.1). An increase of both valgus and varus laxity from mid-flexion to maximal-flexion was seen. A maximum increase of both valgus and varus laxity was found at 110° of flexion. (Fig. 4F). Implantation of the radial head prosthesis results in improved valgus and varus laxity when compared to the unlinked prosthesis with the radial head excised (Fig. 4E). Valgus laxity improves in mid flexion and maximal flexion (P = 0.629). Varus laxity improves near complete extension (P = 0.390). Both instability in the direction of valgus and varus are restored to the amount of instability that was found with the native radial preserved (Table 1).
5.7) 6.1) 5.3) 4.3) 3.8) 4.1) 5.4) 7.5) 7.3) 5.9) 4.6) 4.2) 4.7) 5.9) 6.9) 7.6) 7.5) 7.3) 7.4) 8.4) 10.6) 8.9) 9.4) 8.8) 8.3) 8.4) 9.3) 11.0) 6.9) 7.4) 7.5) 7.6) 8.0) 8.7) 10.1)
This study shows that the stability of the elbow after implantation of the unlinked Latitude total elbow prosthesis is altered when compared to the native elbow. In all tested conditions both the valgus and varus laxity increase from mid to maximal flexion when compared to the native elbow after surgical approach. Maximal increase in both valgus and varus laxity is seen near full flexion in the condition after resection of the radial head. In this condition maximal increase in valgus laxity is approximately 7° and maximal increase in varus laxity is approximately 4° when compared to the condition after the surgical exposure. Crucial in the placement of the Latitude prosthesis is the reproduction of the center of rotation axis (Gramstad et al., 2005). For this purpose meticulous exposure of all ligamentous and bony landmarks is essential. This was achieved by performing an osteotomy of the medial epicondyle and a bony release of the annular ligament. The medial epicondylar osteotomy allowed optimal and anatomical re-fixation of the stabilizing structures on the medial side of the elbow. Optimal fixation of the medial epicondyle was achieved with the trans-osseous sutures in all specimens. The stability tests that were performed before and after surgical exposure showed that the surgical approach had only minimal clinical influence on the elbow kinematics, although it was found to be statistically significant. The changes in valgus and varus stability of the elbow after implantation of the Latitude prosthesis could therefore be fully contributed to the prosthesis design, e.g. the intrinsic constraint of the prosthesis, and the orientation of the prosthesis in relation to the bony landmarks and the soft-tissue constraints. The medial ligamentous complex is the primary stabilizer of the elbow during valgus loading. The radial head is known to be a secondary stabilizer during valgus loading (Beingessner et al., 2004; Jensen et al., 1999, 2005; Morrey et al., 1991). In this study, excision of the radial head resulted in minimal increase of valgus laxity. A critical
M.L. Wagener et al. / Clinical Biomechanics 28 (2013) 502–508
note must be made that all used elbows had intact ligamentous constraints. Elbows with rheumatoid arthritis or posttraumatic deformities often show insufficiency of the collateral ligaments, resulting in diminished stability of the elbow joint. Since elbow arthroplasty is often performed in patients with rheumatoid arthritis, who often have insufficiency of the ligamentous constraints, it can be expected that the radial head will often contribute more to the stability of the elbow during valgus loads than is demonstrated in this study. Furthermore it must be noticed that the radial head can also contribute more to the stability of the elbow when healing of the medial ligamentous complex is impaired after the surgical exposure, or when ligamentous laxity occurs in time after deterioration of the quality of the medial ligamentous complex. In the native elbow excision of the radial head also results in increased instability during varus loading of the elbow. This increase in varus instability is mainly seen near full extension (Jensen et al., 1999). The increase of varus instability can be explained by slacking of the lateral ligament complex after excision of the radial head. In this study a similar, slight, increase in varus instability is seen after excision of the native radial head near full extension. Implantation of the radial head component results in restoration of the stability of the elbow as seen before excision of the native radial head. Valgus stability is improved from mid-flexion to full-flexion when compared to the condition with the radial head excised. Varus stability is improved near full extension. Moreover, restoration of the correct height of the radial head implant is of utmost importance to obtain stability. The correct height of the radial head component is difficult to determine during surgery (Doornberg et al., 2006; Frank et al., 2009; Glabbeek van et al., 2005). Other unlinked elbow prostheses that have the option of implantation of radial head component are the Wright Sorbie-Questor (Wright Medical Technology, Mississauga, ON, Canada) and the Pritchard-ERS (Depuy, Warsaw, IN). The contribution of a monoblock radial head component to the valgus stability of the Wright Sorbie-Questor unlinked total elbow prosthesis was previously described (Inagaki et al., 2002). Contrary to our study they showed increased valgus stability from mid-flexion to full-extension when the radial head component was implanted. This can most likely be explained by a difference in prosthesis design and intrinsic constraint. The kinematic characteristics of the Pritchard-ERS total elbow prosthesis was described by Ramsey et al. (2003). A similar valgus laxity was found as seen in the Latitude prosthesis. The varus laxity of the Pritchard-ERS with radial head component implanted is on the other hand, considerably greater than the varus laxity found in our study. Excision of the radial head resulted in considerably more increase of both valgus and varus laxity than was found with the Latitude total elbow prosthesis. Other studies, in which the contribution of the radial head component to the stability of total elbow prosthesis, e.g. the Mayo Elbow and the AHSC-Volz elbow prosthesis, have to our knowledge not been performed. In the study of King et al. (1994) the stability of the unlinked Capitellocondylar (Johnson and Johnson Orthopaedics, New Brunswick, New Jersey) total elbow prosthesis was analyzed. Since in this study the varus and valgus laxity were added up and presented as one mean maximum valgus-varus laxity for the seventeen used specimens, it is difficult to compare our data to the results of this study. A similar increase of valgus laxity in the direction of full flexion is seen in the figures as presented in their study. Clinically the valgus and varus instability as seen in this study would be scored as “Moderate instability,” according to the “the Mayo Clinic Performance Index for the Elbow” (Morrey and Adams, 1992). Great care should be taken though when translating these biomechanical testing results into clinical functional characteristics. The subjective stability of the elbow that is experienced is grossly dependent on patient demand and expectations. Furthermore the imitated active muscle loads as applied in this study do provide stability to
507
the elbow, but obviously do not match the activity dependent muscle loads in real life. Further research should be performed whether the unlinked version of the Latitude provides enough stability in elbows with insufficiency of collateral ligaments. It might very well be that in elbows with insufficiency of the medial and/or lateral stabilizing structures the unlinked version of the Latitude does not provide enough stability, and that the patient is better off when a linked version of this prosthesis is implanted. 6. Conclusion In conclusion we can state that the unlinked Latitude total elbow prosthesis provides both valgus and varus stability in elbows with intact ligamentous constraints, and that the radial head component only slightly contributes to the stability. Conflict of interest statement No financial remunerations were received by any of the authors. Acknowledgements The authors acknowledge the contributions of M. Barink, PhD, and A. Venner for their support and advice. Financial resources were supplied by Tornier (Stafford, TX) for the purchase of the upper limb specimens, the laboratorial costs, and the costs for the statistical support. The elbow prostheses used in the study were supplied by Tornier. Tornier had no influence on the study design, the collection, analysis, and interpretation of data, or in the writing of the manuscript. References An, K.N., Jacobsen, M.C., Berglund, L.J., Chao, E.Y., 1988. Application of a magnetic tracking device to kinesiologic studies. J. Biomech. 21, 613–620. Beingessner, D.M., Dunning, C.E., Gordon, K.D., Johnson, J.A., King, G.J., 2004. The effect of radial head excision and arthroplasty on elbow kinematics and stability. J. Bone Joint Surg. Am. 86-A, 1730–1739. Dee, R., Sweetnam, D.R., 1970. Total replacement arthroplasty of the elbow joint for rheumatoid arthritis: two cases. Proc. R. Soc. Med. 63, 653–655. Doornberg, J.N., Linzel, D.S., Zurakowski, D., Ring, D., 2006. Reference points for radial head prosthesis size. J. Hand Surg. Am. 31, 53–57. Frank, S.G., Grewal, R., Johnson, J., Faber, K.J., King, G.J., Athwal, G.S., 2009. Determination of correct implant size in radial head arthroplasty to avoid overlengthening. J. Bone Joint Surg. Am. 91, 1738–1746. Glabbeek van, F., Riet van, R.P., Baumfeld, J.A., Neale, P., O'Driscoll, S.W., Morrey, B.F., 2005. The kinematic importance of radial neck length in radial head placement. Med. Eng. Phys. 27, 336–342. Gramstad, G.D., King, G.J., O'Driscoll, S.W., Yamaguchi, K., 2005. Elbow arthroplasty using a convertible implant. Tech. Hand Up Extrem. Surg. 9, 153–163. Inagaki, K., O'Driscoll, S.W., Neale, P.G., Uchiyama, E., Morrey, B.F., An, K.N., 2002. Importance of a radial head component in Sorbie unlinked total elbow arthroplasty. Clin. Orthop. Relat. Res. 123–131. Jensen, S.L., Olsen, B.S., Sojbjerg, J.O., 1999. Elbow joint kinematics after excision of the radial head. J. Shoulder Elbow Surg. 8, 238–241. Jensen, S.L., Olsen, B.S., Tyrdal, S., Sojbjerg, J.O., Sneppen, O., 2005. Elbow joint laxity after experimental radial head excision and lateral collateral ligament rupture: efficacy of prosthetic replacement and ligament repair. J. Shoulder Elbow Surg. 14, 78–84. Kamineni, S., O'Driscoll, S.W., Urban, M., Garg, A., Berglund, L.J., Morrey, B.F., An, K.N., 2005. Intrinsic constraint of unlinked total elbow replacements—the ulnotrochlear joint. J. Bone Joint Surg. Am. 87, 2019–2027. King, G.J., Glauser, S.J., Westreich, A., Morrey, B.F., An, K.N., 1993. In vitro stability of an unconstrained total elbow prosthesis. Influence of axial loading and joint flexion angle. J. Arthroplasty 8, 291–298. King, G.J., Itoi, E., Niebur, G.L., Morrey, B.F., An, K.N., 1994. Motion and laxity of the capitellocondylar total elbow prosthesis. J. Bone Joint Surg. Am. 76, 1000–1008. Morrey, B.F., Adams, R.A., 1992. Semiconstrained arthroplasty for the treatment of rheumatoid arthritis of the elbow. J. Bone Joint Surg. Am. 74, 479–490. Morrey, B.F., An, K.N., 2005. Stability of the elbow: osseous constraints. J. Shoulder Elbow Surg. 14, 174S–178S. Morrey, B.F., An, K.N., Stormont, T.J., 1988. Force transmission through the radial head. J. Bone Joint Surg. Am. 70, 250–256. Morrey, B.F., Tanaka, S., An, K.N., 1991. Valgus stability of the elbow. A definition of primary and secondary constraints. Clin. Orthop. Relat. Res. 187–195.
508
M.L. Wagener et al. / Clinical Biomechanics 28 (2013) 502–508
O'Driscoll, S.W., An, K.N., Korinek, S., Morrey, B.F., 1992. Kinematics of semi-constrained total elbow arthroplasty. J. Bone Joint Surg. Br. 74, 297–299. Olsen, B.S., Sojbjerg, J.O., Dalstra, M., Sneppen, O., 1996. Kinematics of the lateral ligamentous constraints of the elbow joint. J. Shoulder Elbow Surg. 5, 333–341. Ramsey, M., Neale, P., Morrey, B.F., O'Driscoll, S.W., An, K.N., 2003. Kinematics and functional characteristics of the Pritchard ERS unlinked total elbow arthroplasty. J. Shoulder Elbow Surg. 12, 385–390.
Stanley, J., Penn, D., Wasseem, M., 2006. Exposure of the head of the radius using the Wrightington approach. J. Bone Joint Surg. Br. 88, 1178–1182. Szekeres, M., King, G.J., 2006. Total elbow arthroplasty. J. Hand Ther. 19, 245–253. Verbeke, G., Molenberghs, G., 1997. Linear Mixed Models in Practice: An SAS-Oriented Approach. Springer, New York.